Models / Anthropic
Claude Mythos 5
Claude Mythos · Released 2026-06-09
Same underlying model as Claude Fable 5 with safeguards lifted for trusted cyberdefense and research partners. Project Glasswing and a broader trusted-access program — not general consumer availability. $10 / $50 per 1M tokens.
Restricted Mythos-tier access for cyber defenders, critical infrastructure, and approved biology researchers — not a default deployment model.
Per-benchmark positions
Where Claude Mythos 5 places on each public benchmark source that publishes a row. Rank counts every model with a latest row in the same test — not a universal quality score.
No benchmark rows yet for Claude Mythos 5.
Strongest rows in each family
The three highest-scoring carded models in each capability family. Where Claude Mythos 5 shows up, it's highlighted.
- 191
- 291
- 391
- 195
- 294
- 394
- 199
- 297
- 396
- 196
- 296
- 396
- 1100
- 2100
- 3100
- 172
- 272
- 371
- 175
- 275
- 374
- 189
- 287
- 387
- 192
- 292
- 391
- 195
- 292
- 391
Best score per family
A high composite that hides a weak family is a trap. These bars surface the families where this model hasn't been publicly tested, and where it leads.
Where Claude Mythos 5 sits on the scatter
Hover or tab any dot for name, score, and input price.
Full benchmark ledger
Every catalog benchmark for Claude Mythos 5. Scores link to the original source; gaps mean no public row exists yet.
0 of 32 catalog benchmarks have a sourced row for Claude Mythos 5.
Where the numbers come from
- Anthropic announcement· checked 2026-06-09